select records by group with idenitcal variable values in the group

[dr2014]

I need to select obervations which are groupd by variable 'id', where the values for the 2nd variable 'Assignment' is identical i.e 'Z' at all instances of the 'date variable for the id i.e id=2. Please see table below of the original dataset and final dataset needed. Any suggestions on how I can go about this? Please advise.

Original_dataset
ID	Assignment	Date
1	Z	date1
1	D	date2
1	B	date3
1	B	date4
2	Z	date1
2	Z	date2
2	Z	date5
3	Z	date2
3	M	date4
3	Z	date6
Final dataset
ID	Assignment	Date
2	Z	date1
2	Z	date2
2	Z	date5

[ballardw]

I'm not seeing and grouping going on just selection.

Data want;

   set have;

    where Id=2 and assignment='Z';

run;

That may be Id='2' above, I cannot tell whether your Id variable is numeric or character.

[ChrisNZ]

Like this?

proc sql;
select *    
from HAVE
group by ID
having count(unique ASSIGNMENT)=1;
quit;

[mkeintz · Accepted solution]

i think you want all observations for any ID which only has assignment='Z' (i.e. count of non-z's =0).  If so, then this will work:

data want;
  do notz=0 by 0 until (last.id);
    set have;
    by id ;
    if assignment^='Z' then notz=notz+1;
  end;

  do until (last.id);
    set have;
    by id;
    if notz=0 then output;
  end;
run;

[dr2014]

Thanks @mkeintz    . Sorry I wasn; able to reply earlier. I was running into an issue with my login. I did solve the concern just like you did